I'm an average joe but have worked with some radio engineers, and I get the concept of higher frequency = lower penetration. It was fascinating to learn that even rain could interfere with outdoor radio signals if the rain was falling at a similar frequency.
Even fog is awful for higher frequency RF like microwave. On the other hand lower frequency RF on a cloudy day can reach further than a clear day since it bounces off the clouds (ask any old school ham radio operator). Its a fascinating space to work in.
There are now (well have been for years) microwave links from NYC to Chicago, London to Frankfurt, and even top secret R&D for one across the Atlantic between NYC and London for low latency stock trading. They come with a whopping 80% uptime guarantee and enough fine print to effectively make it 0%, but the big firms pay millions a year for a relatively small amount of bandwidth on it. I'm not sure what frequency they're running, but given each span is between 25 and 50 miles, probably up close to 30ghz. If any one span is affected (there's between 25 and 35 spans depending which provider), the whole link goes down and they switch to fiber backup. Each dropped packet on a clear day is likely a bird that flew through the path, and given how high the power levels are, if it flew too close to the dish, it probably became lunch for some ground predator.
A lot of people don't realize that light in fiber is actually significantly slower than the speed of light since there is no such thing as 100% optically perfect glass, and fiber optics rely on bouncing light which makes the actual distance the light travels much longer than the cable itself. RF comes much closer to speed of light (though it varies with frequency and how dense the air is). For example the best fiber path NYC to Chicago is about 7.5msec one way (and they had to trench under the great lakes to get that), and the best microwave link is about 5ms. That translates to potentially billions of dollars in profit for a trading firm.
In the case of wifi, on one hand higher frequency is better, will stay in your own house better (and your neighbors houses) so less interference for everyone. Less prone to interference from stuff like microwaves and cordless devices, higher bandwidth, etc. May need more APs within your home due to reduced range through walls, but since you'll have plenty of channels to choose from, shouldn't be an issue. Hopefully one of the "seamless roaming" standards will be commonplace in the near future and it will be a great experience.
On the other hand, higher frequency = more potential health risks. I know it is considered a conspiracy theory but there is plenty of peer reviewed scientific research showing the effects of RF on human cells, and the mutations that can lead to cancer. I'm glad that Verizon selected sites quite far from my house for their MMwave antennas in my town. But only a matter of time before they're on every pole.
If a microwave running at about 2.5ghz can cook food from the inside at a couple hundred watts (that would be a small microwave, but still gets the job done), then 0.1 watts 24x7x365 just takes longer to cook us.
The big wireless companies did a great job of discrediting those studies a couple decades ago and their contributions to campaigns helps keep them buried.
Free space optics was actually extremely promising (about as close as you can get to actual speed of light outside of a vacuum) but it is so sensitive to anything other than a crystal clear line of site, it just isn't practical. At 1 mile even just high humidity can cause enough refraction to drop the connection. They continue to experiment with it using different wavelengths over long distances, and it is in use on short intra-city connections from building to building. It is actually most useful in outer space, very small amount of power can transmit data very long distances at very high data rates. So 2 objects in the same stable orbit could benefit quite a lot from it. Or even one object in stable orbit like the space station communicating with a static base on Mars etc. Put in some really accurate computer adjusted motors, then one or both objects no longer need to be stable.
I ran a pilot of it between two buildings in Boston. Very cool technology but every time the big screen we had monitoring it showed a dropped packet someone would yell out "bird!".
/Tangent
). But man, it's tempting when I see my poor SSID competing with all those other networks on either side of that nice, blank DFS canvas.
